Behavioural pharmacology

Agmatine attenuates chronic unpredictable mild stress inducedbehavioral alteration in mice

Brijesh G. Taksande, Dharmesh S. Faldu, Madhura P. Dixit, Jay N. Sakaria,Manish M. Aglawe, Milind J. Umekar, Nandkishor R. Kotagale n

Division of Neuroscience, Department of Pharmacology, Shrimati Kishoritai Bhoyar College of Pharmacy, New Kamptee, Nagpur 441002, Maharashtra, India

a r t i c l e i n f o

Article history:Received 2 May 2013Received in revised form19 October 2013Accepted 23 October 2013Available online 29 October 2013

Keywords:AgmatineDepressionAnhedoniaCUMSSplash testFST

a b s t r a c t

Chronic stress exposure and resulting dysregulation of the hypothalamic pituitary adrenal axis developssusceptibility to variety of neurological and psychiatric disorders. Agmatine, a putative neurotransmitterhas been reported to be released in response to various stressful stimuli to maintain the homeostasis.Present study investigated the role of agmatine on chronic unpredictable mild stress (CUMS) inducedbehavioral and biochemical alteration in mice. Exposure of mice to CUMS protocol for 28 days resulted indiminished performance in sucrose preference test, splash test, forced swim test and marked elevation inplasma corticosterone levels. Chronic agmatine (5 and10 mg/kg, ip, once daily) treatment started on day-15 and continued till the end of the CUMS protocol significantly increased sucrose preference, improvedself-care and motivational behavior in the splash test and decreased duration of immobility in the forcedswim test. Agmatine treatment also normalized the elevated corticosterone levels and prevented thebody weight changes in chronically stressed animals. The pharmacological effect of agmatine wascomparable to selective serotonin reuptake inhibitor, fluoxetine (10 mg/kg, ip). Results of present studyclearly demonstrated the anti-depressant like effect of agmatine in chronic unpredictable mild stressinduced depression in mice. Thus the development of drugs based on brain agmatinergic modulationmay represent a new potential approach for the treatment of stress related mood disorders likedepression.

& 2013 Elsevier B.V. All rights reserved.

1. Introduction

Chronic stress and the resulting dysregulation of the hypotha-lamic pituitary adrenal (HPA) axis enhance vulnerability to avariety of neurological disorders. Several neurotransmitters/neu-romodulators and their interactions in CNS regulate response tostress (Grønli et al., 2005). However the exact mechanism behinddeleterious effect of chronic stress exposure is yet to be elucidated.Agmatine, a biogenic amine and putative neurotransmitter, hasbeen implicated in stress response and related disorders (Ariciogluet al., 2003). Agmatine is biosynthesized from amino acid L-arginine by arginine decarboxylase, stored in synaptic vesicles,accumulated by uptake, released by depolarization and inactivatedby agmatinase.

Agmatine metabolized to putrescine and guanido-butanoic acidby an enzyme agmatinase and diamine oxidase respectively (Seereview Uzbay, 2012).

Agmatine interact with α2 adrenergic, imidazoline, NMDA recep-tors and possesses nitric oxide synthase inhibitory activity in brain(see review Uzbay, 2012). Although the physiological role of agmatinein normal brain is largely unknown, the exogenous agmatine hasexhibited several intriguing neurally relevant functions of potentialtherapeutic importance. Systemic administration of agmatine reducedneuronal injury produced by focal ischemia, spinal cord injury andhypoxic ischemia (Lu et al., 2006), diminishes chemically and elec-trically induced convulsions (Demehri et al., 2003). It also producesanxiolytic (Lavinsky et al., 2003), antidepressant (Zomkowski et al.,2002), antinociceptive (Önal et al., 2003), anti-convulsant (Bence et al.,2003), anti-inflammatory (Satriano et al., 2001), antiproliferative(Regunathan and Reis, 1997) and neuroprotective effects (Olmoset al., 1999). Agmatine has been proposed as an adjuvant in thetreatment of several chronic pain syndromes (Paszcuk et al., 2007). Inaddition, several stressful pathological conditions like inflammation,ischemia and lipopolysacarides (LPS) injection increased agmatinelevels in brain. On the other hand, simultaneous treatment withexogenous agmatine attenuated repeated immobilization inducedelevated corticosterone levels and glutamate efflux in brain nucleiassociated with modulation of stress response (M.Y. Zhu et al., 2008;M. Zhu et al., 2008). Further, several nuclei of hypothalamus and

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/ejphar

European Journal of Pharmacology

0014-2999/$ - see front matter & 2013 Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.ejphar.2013.10.041

n Corresponding author. Tel.: þ91 7109 288650; fax; þ91 7109287 094.E-mail address: nandukotagale@gmail.com (N.R. Kotagale).

European Journal of Pharmacology 720 (2013) 115–120

pituitary gland have shown abundant agmatine immunoreactivity(Otake et al., 1998) and its co-localization with neuropeptides(Gorbatyuk et al., 2001). Several studies have reported the neuropro-tective effects of agmatine against cell damage caused by glucocorti-coids and glutamate in primary neuronal cultures of the hippocampus(Iyo et al., 2006; Wang et al., 2006; M.Y. Zhu et al., 2008; M. Zhu et al.,2008). These studies have focused agmatine as an endogenousneuromodulator of stress and suggested that agmatine homeostasismay play an important role in modulation of stress. Endogenouslevels of agmatine increased during stressful conditions in com-pensatory manner, however not high enough to modulate theharmful effect of stressor or inflammation (M.Y. Zhu et al., 2008; M.Zhu et al., 2008). Hence, exogenous administration which restoredthe agmatine levels can exhibit anti-inflammatory, anti-proliferative and neuroprotective effects in rodent. Therefore,one might postulate that as one element of self-protectionmechanisms in the brain, agmatine synthesis is trigged by stressthrough activation of ADC expression, which in turn increasesendogenous agmatine levels as an initial protective response tostress. However, the effect of exogenous agmatine on stress induceddepression and related behavioral alteration remained unexplored.This study was therefore designed to evaluate the role of agmatine inchronic unpredictable mild stress-induced neurobehavioral effectsand biochemical changes in mice.

Chronic unpredictable mild stress (CUMS) exposed animalexhibits several neurobehavioral alteration, resembling the symp-toms of chronic human depression and widely employed forpreclinical screening of antidepressants. In the present study weexamined the effect of chronic agmatine treatment on CUMSinduced anhedonia, motivational behavior, despair, body weightchanges and elevated corticosterone levels in mice.

2. Material and methods

2.1. Subjects

Adult albino Swiss mice of either sex weighing 25–30 g werehoused in polypropylene cages in a temperature (2572 1C),relative humidity (50–70%) and maintained on a 12:12 h light/dark cycle (lights on 07:00–19:00 h). Food and water wereprovided ad libitum except during specific experimental protocols.Control, stressed and treatment group were tested in identicalenvironmental conditions. Cages were changed once per week andwater bottlers were changed three times per week. All experi-mental procedures were carried out under strict compliance withInstitutional Animal Ethical Committee according to guidelines ofthe Committee for the Purpose of Control and Supervision ofExperiments on Animals (CPCSEA), Ministry of Environment andForests; Government of India; New Delhi.

2.2. Drug solutions and administrations

Agmatine sulfate was purchased from Sigma-Aldrich, USAwhile fluoxetine was received as a gift sample from SUN pharma-ceuticals, Baroda, India. Both the drugs were dissolved in salinejust before the experiments and administered intraperitoneally(ip) in a volume of 1 ml/kg. Normal saline (0.9% w/v NaCl) wasused as control.

2.3. Chronic unpredictable mild stress

The chronic unpredictable mild stress (CUMS) protocol wasdesigned to maximize the unpredictable nature of stressor. Restrainedstress, continuous overnight lighting, tail pinch, shaker stress, tilt-cage

and overnight soiled cage were applied once a week over a period of4 weeks. Immediately after completion of each stress session, theanimals were returned to their home cage and transferred to standardlaboratory condition. The duration of CUMS and the nature of differentstressor are based on the available literature (Ducottet and Belzung,2004; Ducottet et al., 2003; Katz et al., 1981; Lu et al., 2006; Willneret al., 1987). Separate group of mice subjected to CUMS were injectedwith saline (1 ml/kg, ip) or fluoxetine (10 mg/kg, ip) or agmatine (2.5,5, and 10mg/kg, ip) from day-15 onwards in between 09:00 and 10:00daily. Unstressed group was injected with saline and handled daily butwas not subjected to any stressors.

In order to avoid the immediate effect of stressor and pharma-cological treatment all the behavioral test were carried out 24 hafter the last stressor except sucrose preference test (SPT). SPT wascarried out on days 0, 14, 21 and 28. The body weight of animalswas measured daily. Separate group of animal was used for everybehavioral paradigm.

2.3.1. Sucrose preference test (SPT)Sucrose preference test was performed as described earlier

(Jindal et al., 2013; Liu et al., 2013) with some modifications. Asagmatine enhanced caloric intake in animals (Taksande et al.,2011) and sucrose consumption can be altered by food and waterdeprivation (Matthews et al., 1995), we did not apply food andwater deprivation in CUMS protocol. SPT was conducted on day 0,14, 21 and 28 of CUMS protocol. Briefly, 72 h before the test, micewere habituated to drink 1% sucrose solution (w/v) and subse-quently exposed to two bottles (1% sucrose solution in one bottleand tap water in another bottle). After habituation mice werehoused in individual cages. Mice were permitted ad libitum accessto 100 ml of sucrose solution (1% w/v) and 100 ml of tap water. Wehave also minimized the effect of side preference in drinkingbehavior by switching the bottle in the middle of the test(Strekolova and Steinbusch, 2010). In addition, we have monitoredthe sucrose preference for 3 h to minimized the % error commonlyobserved with little amount of sucrose consumed in 1 h sucrosepreference test. After 3 h of exposure the consumption volumes ofsucrose solution and tap water were recorded and the sucrosepreference was calculated as [(sucrose consumption)/(water con-sumptionþsucrose consumption)]�100. Mice showing basalsucrose preference less than 60% were considered anhedonic andnot used in the study.

2.3.2. Splash testThe splash test is pharmacological validated animal model

exhibiting motivational behavior of animals. CUMS decreasesgrooming behavior in splash test and this phenomenon is con-sidered to be similar with apathy observed in depressive patients.Animals were isolated in home cages for 24 h and 10% sucrosesolution is squirted on the dorsal coat of mice and the latency toinitiate a grooming behavior as well as the frequency of groomingwas recorded by independent observers unaware of the treatmentfor a period of 5 min where grooming (nose/face grooming, headwashing and body grooming) latency is considered as self-care andgrooming frequency is considered as motivational behavior.

2.3.3. Forced swim testThe procedure was quite similar as described by Porsolt et al.,

(1977) with some modification. Mice were introduced individuallyin plexiglas cylinder (21 cm height�12 cm internal diameter)containing fresh water till a height of 9 cm at 2571 1C and forcedto swim for a 6 min test session. The immobility time wasmeasured by trained observer blind to the treatment. A mousewas considered immobile when it remained floating motionless in

B.G. Taksande et al. / European Journal of Pharmacology 720 (2013) 115–120116

water except making any necessary movement to keep its headabove water.

2.3.4. Corticosterone levelsImmediately after the completion of behavioral tests on day-29,

blood was withdrawn from the retro-orbital plexus of the micein the eppendorff tubes previously rinsed with sodium citrate andcentrifuged at 13,000g for 15 min at 4 1C. Separated plasmawas stored at �20 1C for corticosterone estimation. A quaternarygradient HPLC system equipped with Crestpak C18T-5 column andPDA detector (MD2010 plus) (Jasco, Japan) was used for quantifi-cation of plasma corticosterone. The biochemical estimation onHPLC was carried according to the method described byWoodward and Emery (1987) and Sheikh et al. (2007) with minormodifications. Briefly, 50 ml of plasma was extracted with 1 ml ofDCM–Ether mixture (DCM:Ether—50:50) on mechanical shaker for15 min. Supernatant (50 ml) was transferred to eppendorff tubeand evaporated under the slow stream of nitrogen. After completeevaporation, 1 ml of mobile phase (Water:Methanol—80:20) wasadded and 20 ml of this was injected into the HPLC (Flow rate-1 ml/min and estimated at 243 nm).

2.4. Statistical analysis

The data were expressed as mean7S.E.M. The results of splashtest, forced swimming test, open field were analyzed by one-wayANOVA followed by the Newman–Keul test. Sucrose preferenceand body weight changes were analyzed by two-way ANNOVAwith post-hoc Bonferroni mean comparisons. Corticosteronelevels were analyzed by unpaired t test. Results of statisticaltests with Po0.05 were considered significant.

3. Results

3.1. Effect of chronic administration of agmatine in stress inducedanhedonia

Fig. 1 shows the effect of chronic unpredictable mild stress onsucrose preference test. Chronic exposure of animal to unpredict-able mild stress significantly decrease the sucrose preference atday-14 (Po0.001) and the effect was continued upto day-28(Po0.001) as compared to unstressed animals [FStress� Time (3,30)¼75.59, Po0.001; FStress(1, 30)¼3775.72, Po0.001; FTime (3,30)¼18.88, Po0.001- Two way ANOVA]. The sucrose preferenceon day-28 was significantly lower as compared to day-14(Po0.001). Post-hoc Bonferroni mean comparisons revealed sig-nificant difference in sucrose preference between the groups onday-14 (Po0.001) and day-28 (Po0.001).

Agmatine (5 and 10 mg/kg) [FTreatment� Time (12, 75)¼11.66,Po0.001; FTreatment (4, 75)¼62.71, Po0.001; FTime (3, 75)¼35.35, Po0.001] and fluoxetine (10 mg/kg) [FTreatment� Time (6,45)¼20.94, Po0.001; FTreatment (2, 45)¼174.63, Po0.001; FTime

(3, 45)¼9.72, Po0.001] treatment from day-15 onwards comple-tely reversed the anhedonia induced by CUMS paradigm on day-28. Post-hoc Bonferroni mean comparisons revealed that thesucrose preference in the stressed animals treated with agmatine(5 and 10 mg/kg) and fluoxetine (10 mg/kg, ip) were significantlydifferent as compared to stressed mice treated with vehicle onday-28. However no significant effect of treatment was observedon day-21. The difference in the % sucrose preference on day 21and 28 in the animal treated with agmatine (10 mg/kg) was notstatistically significant as compared to fluoxetine (10 mg/kg, ip)treatment. Administration of agmatine and fluoxetine to non-stressed group did not produce any effect on sucrose preference

test as compared to vehicle treated non-stressed group (data notshown).

3.2. Effect of agmatine in stress induced alterationsin self-care and motivational behavior

Chronic agmatine (5 and 10 mg/kg) and fluoxetine (10 mg/kg)treatment from third week onward of CUMS protocol restored themotivational and self-care behavior in the splash test (Fig. 2). Oneway ANOVA followed by post-hoc Newman–Keuls comparisonsdemonstrated that agmatine [5(Po0.001 and Po0.001) and10 mg/kg (Po0.001 and Po0.001) respectively] significantlydecreased the grooming latency [F (4, 29)¼22.08, Po0.001][Fig. 2A] and increased grooming frequency [Fig. 2B] as comparedto vehicle treated stressed animals. Similarly fluoxetine (10 mg/kg)[(Po0.001and Po0.001) respectively] treatment to separategroup of animals showed significant decrease in grooming latency[F (2, 17)¼94.95, Po0.001] and increase in grooming frequency[F (2, 17)¼89.33, Po0.001]. Agmatine (2.5–10 mg/kg) and fluox-etine (10 mg/kg) treatment to non-stressed mice did not affect thebehavior in splash test (data not shown).

3.3. Effect of agmatine in FST

As shown in Fig. 3, CUMS exposure significantly increasesimmobility duration in stressed animals as compared to non-stressed group (Po0.001). Repeated administration of agmatine(5 and 10 mg) dose dependently reduces the % immobility time inmice exposed to CUMS by 23% (Po0.05) and 42% (Po0.001)respectively as compared to vehicle treated stressed animal [F (4,29)¼10.08, Po0.001]. Agmatine at the doses used here did notinfluence the immobility duration in non-stressed mice (data notshown). Fluoxetine treatment at dose of 10 mg/kg to a separategroup of mice also reduced immobility time in CUMS exposedanimal by 46% (Po0.001) [F (2, 17)¼51.51, Po0.001].

Fig. 1. Effects of agmatine and fluoxetine treatment on sucrose preference test.Mice (n¼6) were subjected to chronic unpredictable mild stress (CUMS) andinjected with saline (1 ml/kg, ip) or fluoxetine (10 mg/kg, ip) or agmatine (2.5–10 mg/kg, ip) from day-15 onwards and sucrose preference was evaluated on days0, 14, 21 and 28. Control group received saline (1 ml/kg, ip) under the sameschedule and was not subjected to stress. Each point indicate mean sucrosepreference (%)7S.E.M. $Po0.001 vs. saline treated non-stressed animals;nPo0.05 nnPo0.001 vs. saline treated CUMS animals (two way ANOVA post-hocBonferroni mean comparisons).

B.G. Taksande et al. / European Journal of Pharmacology 720 (2013) 115–120 117

3.4. Effect of agmatine on % body weight change in CUMS induceddepression

Body weight was monitored daily during the experimentalprotocol. Two-way ANOVA showed a significant reduction in thebody weight of the CUMS exposed animals as compared to non-stressed mice [FStress� Time (3, 30)¼11.98, Po0.001; FStress (1, 30)¼44.91, Po0.001; FTime (3, 30)¼6.59, Po0.01]. Post hoc analysis byBonferroni multiple comparison showed significant difference in %body weight change on day-21 (Po0.01) and 28 (Po0.05). Nosignificant change in the body weight was observed in the micethat received repeated agmatine treatment and subjected toCUMS when compared against saline treated non-stressedanimals [FTreatment�Time (15, 90)¼0.86, P¼0.61; FTreatment (5,90)¼7.46, Po0.001; FTime (3, 90)¼1.49, P¼0.22]. As depicted inFig. 4, fluoxetine (10 mg/kg) significantly prevented reduction inbody weight in animals subjected to CUMS protocol. Two wayANOVA indicated insignificant change in the body weight influoxetine (10 mg/kg) treated animals as compared to salineinjected mice exposed to CUMS [FTreatment� Time (6, 45)¼2.84,Po0.05; FTreatment (2, 45)¼25.23, Po0.001; FTime (3, 45)¼0.57,P¼0.64]. No significant difference was determined between thestressed animals injected with agmatine (10 mg/kg) and fluox-etine (10 mg/kg) (post-hoc Bonferroni mean comparisons).

3.5. Agmatine attenuates CUMS induced increased plasmacorticosterone levels in mice

HPLC analysis demonstrated the significant higher levels ofplasma corticosterone in the animals subjected to CUMS for theperiod of 28 days (Po0.001) as compared to non-stressed controlgroup (t¼13.93; df¼4, unpaired t test). As depicted in Fig. 5,plasma corticosterone levels were significantly lower and compar-able to the non-stressed group in agmatine [5 (Po0.01) and10 mg/kg (Po0.001)] (F (4, 14)¼30.63, (Po0.001)) and fluoxetine(Po0.001) [F (2, 8)¼74.75, (Po0.001)] treated mice.

Fig. 2. Effects of agmatine and fluoxetine treatment on motivational behavior insplash test. Mice (n¼6) were subjected to chronic unpredictable mild stress(CUMS) and injected with saline (1 ml/kg, ip) or fluoxetine (10 mg/kg, ip) oragmatine (2.5–10 mg/kg, ip) from day-15 onwards and grooming latency [(seconds(s))] (Fig. 2A) and frequency (Fig. 2B) were evaluated 24 h after the last stressor.Control group received saline (1 ml/kg, ip) under the same schedule and was notsubjected to stress. Each bar indicate mean grooming latency (s)/groomingfrequency7S.E.M. $Po0.01, $$Po0.001 vs. saline treated non-stressed animals;nPo0.001 vs. saline treated CUMS animals (one way ANOVA post-hoc Newman–Keul mean comparisons).

Fig. 3. Effects of agmatine and fluoxetine treatment on immobility time in forcedswim test (FST). Mice (n¼6) were subjected to chronic unpredictable mild stress(CUMS) and injected with saline (1 ml/kg, ip) or fluoxetine (10 mg/kg, ip) oragmatine (2.5–10 mg/kg, ip) from day-15 onwards and immobility time [(seconds (s))]was determined 24 h after the last stressor. Control group received saline (1 ml/kg, ip)under the same schedule and was not subjected to stress. Each bar indicate meanimmobility time (s)7S.E.M. $Po0.001vs saline treated non-stressed animals;nPo0.05, nn Po0.001 vs. saline treated CUMS animals (one way ANOVA post-hocNewman–Keul mean comparisons).

Fig. 4. Effects of agmatine and fluoxetine treatment on changes in body weight.Mice (n¼6) were subjected to chronic unpredictable mild stress (CUMS) andinjected with saline (1 ml/kg, ip) or fluoxetine (10 mg/kg, ip) or agmatine (2.5–10mg/kg, ip) from day-15 onwards and body weights were recorded on day 0, 14, 21and 28. Control group received saline (1 ml/kg, ip) under the same schedule andwas not subjected to stress. Each point indicate mean changes in body weight (%)7S.E.M. nPo0.05, nnPo0.01 vs. saline injected non-stressed animals (two wayANOVA post-hoc Bonferroni mean comparisons).

B.G. Taksande et al. / European Journal of Pharmacology 720 (2013) 115–120118

4. Discussion

Stressful life experiences are important etiological factors in theonset and maintenance of depressive episode (Lee et al., 2002). Inview of this, an animal model of CUMS induced depression has beendeveloped to replicate the pathogenesis of depressive illness.Number of experimental evidence suggests that CUMS exposureto animals induce alteration in behavioral and biochemical para-meters resembling symptoms of clinical depression (Luo et al.,2008). Consistent with several earlier findings, the results of thepresent study showed that mice when subjected to chronic expo-sure to a variety of mild stressor regimen induced significantreduction in sucrose preference test, grooming behavior in splashtest, body weights and increased immobility in forced swim testand plasma corticosterone level as compared to non-stressed mice.

Sucrose preference test is an indicator of anhedonia like beha-vioral change (Willner, 2005). The results of the present studyshowed that the long term treatment of agmatine significantlysuppressed anhedonia induced by CUMS protocol suggesting thepotential antidepressant like effect of agmatine. The antidepressanteffect was specific to stress exposure as agmatine did not influencethe sucrose consumption in normal non-stressed group. Similarly,agmatine treatment did not influence the water intake in naïveanimals. The anti-depressant like effect of agmatine was com-pared to selective serotonin reuptake inhibitor—fluoxetine withthe only difference that agmatine has shown delayed onset ofaction. The sucrose consumption was normalized by fluoxetinetreatment on day-21 whereas it was evident on day-28 inagmatine treated group. This could be attributed to lower dosesof agmatine used in the present study and their differentialmechanism of action. Further, intraperitoneally injected agma-tine extensively metabolized peripherally in liver and kidney andhas very short biological half life (Halaris and Plietz, 2007). Thedelayed onset of action and significant weight gain is a possibleside effect with SSRI and other antidepressant including tricyclicantidepressant and MAO inhibitors (Ruetsch et al., 2005). In ourstudy, although agmatine exhibited delayed onset of action ascompared to fluoxetine, it did not induced significant changes inbody weight of animal after chronic administration. Moreover,unlike other antidepressant, it did not affect the locomotorcounts of animal. Thus agmatinergic modulation may offer

several advantages and may be useful in combination with otherantidepressant in the treatment of resistant depression.

The splash test is direct measure of self-care and motivationalbehavior. Chronic stress exposure to mice resulted in increase inlatency and decreased frequency of grooming behavior, indicatingloss of self-care and motivational behavior in mice subjected toCUMS procedure. The disturbance in grooming behavior is con-sidered to mimic apathy observed in clinical depression (Willner,2005). Agmatine and fluoxetine treatment prevented the effect ofCUMS on splash test. Indeed, similar effects have been alreadyreported for fluoxetine and CRF II antagonist (Ducottet et al., 2003;Santarelli et al., 2003). Chronic stress has been shown to drama-tically increase the immobility time of mice in FST (Zhou et al.,2007). Similarly our CUMS protocol also increased immobility timein forced swim test in mice. Chronic treatment with agmatine andfluoxetine attenuated the CUMS associated reduction in immobi-lity time. These behavioral observations demonstrated that agma-tine produced an antidepressant like effect in chronically stressedmice. In parallel with our observation, simultaneous treatmentwith exogenous agmatine attenuated repeated immobilizationinduced architectural alteration in hippocampus and prefrontalcortex such as elevated plasma corticosterone levels and highglutamate influx (M.Y. Zhu et al., 2008; M. Zhu et.al., 2008).Agmatine also attenuated stress and lipopolysaccharide inducedhyperthermia in rats (Aricioglu and Regunathan, 2005). MoreoverBernstein et al. (2012) have shown up-regulation of the agma-tine degrading enzyme agmatinase in patients with unipolarand bipolar depression suggesting that reduction of endogen-ous brain agmatine level may play a central role in thepathogenesis of depression. Thus, it can be inferred from ourresults and available literature that endogenous agmatine systemmay play an important role in adaptation response to chronicstress in order to maintain brain homeostasis.

In the present study, the CUMS exposure led to significantreduction in body weights compared to saline treated nonstressedanimals. The results are in agreement with previous data showingthat chronic mild stress lead to reduced body weight in exposedanimals (Wang et al., 2008). Fluoxetine and agmatine treatmentsignificantly prevented the body weight reduction induced byCUMS protocol. It is important to note that, body weight reductionnot only provides face validity to CUMS procedure but also consideras a diagnostic criterion for a major depressive episode in DSM-IV.Thus agmatine treatment might offer additional advantages in thetreatment of stress related disorders associated with anorexia andsignificant loss of body weights. However, body weight changes canalso be considered as a confounding variable in sucrose consump-tion and could lead to false results in CUMS protocol. Our results ofanhedonia should not be attributed to body weight changes asabsolute sucrose consumption may be influenced by body weightgain, sucrose preference may not be influenced and consider as abetter index of anhedonia (Matthews et al., 1995).

HPA axis plays a critical role in eliciting physiological responsesto various stressful stimuli (Pan et al., 2006). Acute stress activatesHPA axis with simultaneous increase in levels of corticosterone inrodents or cortisol in human being. Sustained activation of HPAaxis is associated with an abnormally high blood glucocorticoidlevel, which may eventually lead to pathological conditions suchas depression (Johnson et al., 2006). Thus, the normalization ofHPA axis and glucocorticoid levels may be critically involved in thetherapeutic action of antidepressant drug. Normalization of theHPA system has been shown to be a prerequisite for stableremission of the disease (Holsboer, 2000), which has been shownto occur during successful antidepressant treatment (Barden et al.,1995), again supporting the importance of the stress hormonesystem in the development and maintenance of affective disor-ders. In accordance with several earlier findings, we found that

Fig. 5. Effects of agmatine and fluoxetine treatment on plasma corticosteronelevels. Mice (n¼3) were subjected to chronic unpredictable mild stress (CUMS)and injected with saline (1 ml/kg, ip) or fluoxetine (10 mg/kg, ip) or agmatine(2.5–10 mg/kg, ip) from day-15 onwards. 24 h after the last stressor and bloodsamples were withdrawn to determine corticosterone level. Control group receivedsaline (1 ml/kg, ip) under the same schedule and was not subjected to stress. Eachbar indicate mean plasma corticosterone (ng/ml)7S.E.M. $Po0.001vs salinetreated non-stressed animals; nPo0.01, nnPo0.001 vs. saline treated CUMSanimals (one way ANOVA posthoc Newman Keul mean comparisons).

B.G. Taksande et al. / European Journal of Pharmacology 720 (2013) 115–120 119

CUMS significantly elevated corticosterone levels in mice plasma,and chronic agmatine treatment prevented these alterations. Thuspreventive effects of agmatine on CUMS associated behavioralalterations might be linked with reduction in the corticosteronelevels. Although we have not investigated specific mechanism,target receptor and enzymes like α2 adrenoceptor, imidazolinereceptor, NMDA receptor or nitric oxide might be involved inantidepressant like effect of agmatine. Further more specificstudies are required for better understanding of this proposedmechanism.

To conclude, the results of present study demonstrated anti-depressant like effect of agmatine in chronic unpredictable mildstress induced depression in mice. Chronic agmatine (5 and 10 mg/kg, ip, once daily) treatment started on day-15 and continued tillthe end of the CUMS protocol significantly increased sucrosepreference, improved self-care and motivational behavior in splashtest, decreased duration of immobility in forced swim test and alsonormalized the body weight changes and elevated corticosteroneplasma levels in mice. Thus the development of drugs based onmodulation of agmatinergic system in brain may represent a newpotential approach for the treatment of stress related neurologicaldisease like depression.

Appendix A. Supplementary material

Supplementary data associated with this article can be found inthe online version at http://dx.doi.org/10.1016/j.ejphar.2013.10.041.

References

Aricioglu, F., Regunathan, S., 2005. Agmatine attenuates stress- and lipopolysaccharide-induced fever in rats. Physiol. Behav. 85, 370–375.

Aricioglu, F., Regunathan, S., Piletz, J., 2003. Is agmatine an endogenous factoragainst stress? Ann. NY. Acad. Sci. 1009, 127–132.

Barden, N., Reul, J.M., Holsboer, F., 1995. Do antidepressants stabilize mood throughactions on the hypothalamic–pituitary–adrenocortical system? Trends Neu-rosci. 18, 6–11.

Bence, K., David, R., Stables, J.P., Crooks, P.A., 2003. An in vivo evaluation of theantiseizure activity and acute neurotoxicity of agmatine. Pharmacol. Biochem.Behav. 74, 771–775.

Bernstein, H.G., Stich, C., Jäger, K., Dobrowolny, H., Wick, M., Steiner, J., Veh, R.,Bogerts, B., Laube, G., 2012. Agmatinase, an inactivator of the putative endogen-ous antidepressant agmatine, is strongly upregulated in hippocampal interneur-ons of subjects with mood disorders. Neuropharmacology 62, 237–246.

Demehri, S., Homayoun, H., Honar, H., Riazi, K., Vafaie, K., Roushanzamir, F.,Dehpour, A.R., 2003. Agmatine exerts anticonvulsant effect in mice: modulationby alpha 2-adrenoceptors and nitric oxide. Neuropharmacology 45, 534–542.

Ducottet, C., Belzung, C., 2004. Behavior in the elevated plus-maze predicts copingafter subchronic mild stress in mice. Physiol. Behav. 81, 417–426.

Ducottet, C., Griebel, G., Belzung, C., 2003. Effects of the selective nonpeptidecorticotropin-releasing factor receptor 1 antagonist antalarmin in the chronicmild stress model of depression in mice. Prog. Neuro-Psychopharmacol. Biol.Psychiatry 27, 625–631.

Gorbatyuk, O.S., Milner, T.A., Wang, G., Regunathan, S., Reis, D.J., 2001. Localizationof agmatine in vasopressin and oxytocin neurons of the rat hypothalamicparaventricular and supraoptic nuclei. Exp. Neurol. 171, 235–245.

Grønli, J., Murison, R., Fiske, E., Bjorvatn, B., Sørensen, E., Portas, C.M., Ursin, R.,2005. Effects of chronic mild stress on sexual behavior, locomotor activity andconsumption of sucrose and saccharine solutions. Physiol. Behav. 84, 571–577.

Halaris, A., Plietz, J., 2007. Agmatine metabolic pathway and spectrum of activity inbrain. CNS Drugs 21 (11), 885–900.

Holsboer, F., 2000. The corticosteroid receptor hypothesis of depression. Neurop-sychopharmacology 23, 477–501.

Iyo, A.H., Zhu, M.Y., Ordway, G.A., 2006. Expression of arginine decarboxylase inbrain regions and neuronal cells. J. Neurochem. 96, 1042–1050.

Jindal, A., Mahesh, R., Bhatt, S., 2013. Etazolate, a phosphodiesterase 4 inhibitorreverses chronic unpredictable mild stress-induced depression-like behaviorand brain oxidative damage. Pharmacol. Biochem. Behav. 105, 63–70.

Johnson, S.A., Fournier, N.M., Kalynchuk, L.E., 2006. Effect of different doses ofcorticosterone on depression-like behavior and HPA axis responses to a novelstressor. Behav. Brain Res. 168, 280–288.

Katz, R.J., Roth, K.A., Carroll, B.J., 1981. Acute and chronic stress effects on open fieldactivity in the rat: implications for a model of depression. Neurosci. Biobehav.Rev. 5, 247–251.

Lavinsky, D., Arteni, N.S., Netto, C.A., 2003. Agmatine induces anxiolysis in theelevated pluse maze task in adults rat. Behav. Brain Res. 141, 19–24.

Lee, A.L., Ogle, W.O., Sapolsky, R.M., 2002. Stress and depression: possible links toneuron death in the hippocampus. Bipolar Disord. 4, 117–128.

Liu, Y., Jia, G., Gou, L., Sun, L., Fu, X., Lan, N., Li, S., Yin, X., 2013. Antidepressant-likeeffects of tea polyphenols on mouse model of chronic unpredictable mildstress. Pharmacol. Biochem. Behav. 104, 27–32.

Lu, X.-Y., Kim, C.S., Frazer, A., Zhang, W., 2006. Leptin: a potential novel anti-depressant. Proc. Natl. Acad. USA 103, 1593–1598.

Luo, D., An, S., Zhang, X., 2008. Involvement of hippocampal serotonin andneuropeptide Y in depression induced by chronic unpredicted mild stress.Brain Res. Bull. 77, 8–12.

Matthews, K., Forbes, N., Reid, I.C., 1995. Sucrose consumption as an hedonic measurefollowing chronic unpredictable mild stress. Physiol. Behav. 57, 241–248.

Olmos, G., DeGregorio-Rocasolano, N., Paz Regalado, M., Gasull, T., AssumpcioBoronat, M., Trullas, R., 1999. Protection by imidazol(ine) drugs and agmatine ofglutamate induced neurotoxicity in cultured cerebellar granule cells throughblockade of NMDA receptor. Br. J. Pharmacol. 127, 1317–1326.

Önal, A., Delen, Y., Ülker, S., Soykan, N., 2003. Agmatine attenuates neuropathic painin rats: possible mediation of nitric oxide and noradrenergic activity in thebrainstem and cerebellum. Life Sci. 73, 413–428.

Otake, K., Ruggiero, D.A., Regunathan, S., Wang, H., Milner, T.A., Reis, D.J., 1998.Regional localization of agmatine in the rat brain: an immunocytochemicalstudy. Brain Res. 787, 1–14.

Paszcuk, A.F., Gadotti, V.M., Tibola, D., Quintão, N.L., Rodrigues, A.L.S., Calixto, J.B.,Santos, A.R., 2007. Anti-hypernociceptive properties of agmatine in persistentinflammatory and neuropathic models of pain in mice. Brain Res. 1159,124–133.

Pan, Y., Zhang, W.-Y., Xia, X., Kong, L.-D., 2006. Effects of icariin on hypothalamic–pituitary–adrenal axis action and cytokine levels in stressed Sprague–Dawleyrats. Biol. Pharm. Bull. 29, 2399–2403.

Porsolt, R.D., Le Pichon, M., Jalfre, M., 1977. Depression: a new animal modelsensitive to antidepressant treatments. Nature 266, 730–732.

Regunathan, S., Reis, D.J., 1997. Stimulation of imidazoline receptors inhibitsproliferation of human coronary artery vascular smooth muscle cells. Hyper-tension 30, 295–300.

Ruetsch, O., Viala, A., Bardou, H., Martin, P., Vacheron, M.N., 2005. Psychotropicdrugs induced weight gain: a review of the literature concerning epidemiolo-gical data, mechanisms and management. Encephale 31 (4), 507–516.

Santarelli, L., Saxe, M., Gross, C., Surget, A., Battaglia, F., Dulawa, S., Weisstaub, N.,Lee, J., Duman, R., Arancio, O., 2003. Requirement of hippocampal neurogenesisfor the behavioral effects of antidepressants. Science 301, 805–809.

Satriano, J., Schwartz, D., Ishizuka, S., Lorie, M.J., Thomson, S.C., Gabbai, F., 2001.Supression of inducible nitric oxide generation by agmatine aldehyde: bene-ficial effects in sepsis. J. Cell. Physiol. 188, 313–320.

Sheikh, N., Ahmad, A., Siripurapu, K., Kuchibhotla, V., Singh, S., Palit, G., 2007. Effectof Bacopa monniera on stress induced changes in plasma corticosterone andbrain monoamines in rats. J. Ethnopharmacol. 111, 671–672.

Strekolova, T., Steinbusch, H.W.M., 2010. Measuring behavior in mice with chronicstress depression paradigm. Prog. Neuropsychopharmacol. Biol. Psychiatry 34,348–351.

Taksande, B., Kotagale, N., Nakhate, K., Mali, P., Kokare, D., Hirani, K., Subhedar, N.,Chopde, C., Ugale, R., 2011. Agmatine in the hypothalamic paraventricularnucleus stimulates feeding in rats: involvement of neuropeptide Y. Br. J.Pharmacol. 164, 704–718.

Uzbay, T.I., 2012. The pharmacological importance of agmatine in the brain.Neurosci. Biobehav. Rev. 36, 502–519.

Wang, W.P., Iyo, A.H., Miguel-Hidalgo, J., Regunathan, S., Zhu, M.Y., 2006. Agmatineprotects against cell damage induced by NMDA and glutamate in culturedhippocampal neurons. Brain Res. 1084 (1), 210–216. (21).

Wang, D., An, S.C., Zhang, X., 2008. Prevention of chronic stress-induced depres-sion-like behaviour by inducible nitric oxide inhibitor. Neurosci. Lett. 433,59–64.

Willner, P., 2005. Chronic mild stress (CMS) revisited: consistency and behavioural–neurobiological concordance in the effects of CMS. Neuropsychobiology 52,90–110.

Willner, P., Towell, A., Sampson, D., Sophokleous, S., Muscat, R., 1987. Reduction ofsucrose preference by chronic unpredictable mild stress, and its restoration by atricyclic antidepressant. Psychopharmacology 93, 358–364.

Woodward, C.J., Emery, P.W., 1987. Determination of plasma corticosterone usinghigh-performance liquid chromatography. J. Chromatogr. 419, 280–284.

Zhou, Q.G., Hu, Y., Hua, Y., Hu, M., Luo, C.X., Han, X., Zhu, X.J., Wang, B., Xu, J.S., Zhu,D.Y., 2007. Neuronal nitric oxide synthase contributes to chronic stress‐induceddepression by suppressing hippocampal neurogenesis. J. Neurochem. 103,1843–1854.

Zhu, M.Y., Wang, W.P., Cai, Z.W., Regunathan, S., Ordway, G., 2008. Exogenousagmatine has neuroprotective effects against restraint‐induced structuralchanges in the rat brain. Eur. J. Neurosci. 27, 1320–1332.

Zhu, M., Wang Huang, J., Feng, Y.Z., Regunathan, S., Bissette, G., 2008. Repeatedimmobilization stress alters rat hippocampal and prefrontal cortical morphologyin parallel with endogenous agmatine and arginine decarboxylase levels.Neurochem. Int. 53, 346–354.

Zomkowski, A.D., Hammes, L., Lin, J., Calixto, J.B., Santos, A.R.S., Rodrigues, A.L.S.,2002. Agmatine produces antidepressant-like effects in two models of depres-sion in mice. Neuroreport 13, 387–391.

B.G. Taksande et al. / European Journal of Pharmacology 720 (2013) 115–120120